import numpy as np
import matplotlib.pyplot as plt
from matplotlib import cm
import networkx as nx
from utils import SIR_net, SIR_net_adaptive
import utils

Testing the effect of rewiring in the information network

N = 1000 # number of nodes
ave_degree = 15
p = ave_degree / N  # edge probability
initial_infecteds = np.random.choice(np.arange(N), size=100, replace=False)
#initial_no_vax = np.arange(100,400)
initial_no_vax = 300


mu = 0.1       # recovery rate
beta = 0.15    # infection rate
pro = 0.      # rate of classical media influence on people
pol = 0.5      # propensity of opinion polarization
p_sym = 0.     # prob for an infected person to have severe symptoms
r = 0.01        # rate of vaccination for PV

G_phys = nx.barabasi_albert_graph(N, int(ave_degree/2))
G_info_static = G_phys.copy()
utils.initNET_SI(G_info_static, I0=initial_no_vax, I_seed=np.arange(10)) # initialize information network
#utils.initNET_rnd(G_info_static, initial_no_vax)
G_info_dynamic = G_info_static.copy()


sim_static = SIR_net_adaptive(G_phys, G_info_static, beta=beta, mu=mu, r=r, pro=pro, pol=pol, p_sym=p_sym, initial_infecteds=initial_infecteds, rewiring=False)
sim_dynamic = SIR_net_adaptive(G_phys, G_info_dynamic, beta=beta, mu=mu, r=r, pro=pro, pol=pol, p_sym=p_sym, initial_infecteds=initial_infecteds)

simulation until time t=74

colors = cm.get_cmap('winter', 2)
fig, ax = plt.subplots(nrows=1,ncols=1,figsize=(10,6))
ax.plot(sim_static[0], sim_static[2], lw=3, c=colors(0), label= "Static information network", alpha=1)
ax.plot(sim_dynamic[0], sim_dynamic[2], lw=3, c=colors(1), label= "Dynamic information network", alpha=1)
ax.legend(fontsize=15)
ax.set_ylabel("Prevalence", size=15)
ax.set_xlabel("time", size=15)
ax.set_xlim(0, 50)

(0.0, 50.0)

print('infoNET clustering coefficients:')
print('static:', nx.algorithms.cluster.average_clustering(G_info_static))
print('dynamic:', nx.algorithms.cluster.average_clustering(G_info_dynamic))
utils.plot_info_network(G_info_static)
utils.plot_info_network(G_info_dynamic)

infoNET clustering coefficients:
static: 0.04886055257342535
dynamic: 0.5566707977857527

vac = 0
rec = 0
for i in G_phys.nodes():
    if G_phys.nodes[i]['inf_status'] == 'V':
        vac += 1
    elif G_phys.nodes[i]['inf_status'] == 'R':
        rec += 1    
print('vaccinated', vac)
print('recovered', rec)

vaccinated 15
recovered 984

Testing if with a faster dynamics we can appreciate differences among curves with this second scheme

N = 1000 # number of nodes
ave_degree = 15
p = ave_degree / N  # edge probability
initial_infecteds = np.random.choice(np.arange(N), size=100, replace=False)

mu = 0.02       # recovery rate
beta = 0.03    # infection rate
pro = 0.0005       # rate of classical media influence on people
pol = 0.5       # propensity of opinion polarization
p_sym = 0.05    # prob for an infected person to have severe symptoms
r = 0.1         # rate of vaccination for PV


G_phys = nx.barabasi_albert_graph(N, int(ave_degree/2))
G_info_0 = G_phys.copy()
G_info_50 = G_phys.copy()
G_info_100 = G_phys.copy()
G_info_300 = G_phys.copy()
utils.initNET_SI(G_info_0, I0=0, I_seed=[])
utils.initNET_SI(G_info_50, I0=50, I_seed=np.arange(10))
utils.initNET_SI(G_info_100, I0=100, I_seed=np.arange(10))
utils.initNET_SI(G_info_300, I0=300, I_seed=np.arange(10))

sim1 = SIR_net(G_phys, beta=beta, mu=mu, initial_infecteds=initial_infecteds)
sim2 = SIR_net_adaptive(G_phys, G_info_0, beta=beta, mu=mu, r=r, pro=pro, pol=pol, p_sym=p_sym, initial_infecteds=initial_infecteds)
sim3 = SIR_net_adaptive(G_phys, G_info_50, beta=beta, mu=mu, r=r, pro=pro, pol=pol, p_sym=p_sym, initial_infecteds=initial_infecteds)
sim4 = SIR_net_adaptive(G_phys, G_info_100, beta=beta, mu=mu, r=r, pro=pro, pol=pol, p_sym=p_sym, initial_infecteds=initial_infecteds)
sim5 = SIR_net_adaptive(G_phys, G_info_300, beta=beta, mu=mu, r=r, pro=pro, pol=pol, p_sym=p_sym, initial_infecteds=initial_infecteds)

simulation until time t=79

colors = cm.get_cmap('winter', 5)
fig, ax = plt.subplots(nrows=1,ncols=1,figsize=(10,6))
ax.plot(sim2[0], sim2[2], lw=3, c=colors(0), label= "0 No-Vax", alpha=1)
ax.plot(sim3[0], sim3[2], lw=3, c=colors(1), label= "50 No-Vax", alpha=1)
ax.plot(sim4[0], sim4[2], lw=3, c=colors(2), label= "100 No-Vax", alpha=1)
ax.plot(sim5[0], sim5[2], lw=3, c=colors(3), label= "300 No-Vax", alpha=1)
ax.plot(sim1[0], sim1[2], lw=3, c=colors(4), label= "SIR", alpha=1)
ax.legend(fontsize=15)
ax.set_ylabel("Prevalence", size=15)
ax.set_xlabel("time", size=15)
ax.set_xlim(0,400)

(0.0, 400.0)

print('total infected with 0 novax:', sim2[4])
print('total infected with 50 novax:', sim3[4])
print('total infected with 100 novax:', sim4[4])
print('total infected with 300 novax:', sim5[4])

total infected with 0 novax: 842
total infected with 50 novax: 884
total infected with 100 novax: 890
total infected with 300 novax: 901

print('infoNET clustering coefficients:')
print('dynamic:', nx.algorithms.cluster.average_clustering(G_info_300))
utils.plot_info_network(G_info_300)

infoNET clustering coefficients:
dynamic: 0.19241202877390212

Simulations with different polarization

N = 1000 # number of nodes
ave_degree = 15
p = ave_degree / N  # edge probability
initial_infecteds = np.random.choice(np.arange(N), size=100, replace=False)

mu = 0.02       # recovery rate
beta = 0.03    # infection rate
pro = 0.0005    # rate of classical media influence on people
pol = np.arange(1,20)/20       # propensity of opinion polarization
p_sym = 0.05    # prob for an infected person to have severe symptoms
r = 0.4         # rate of vaccination for PV

networks = []
inf = []

for i in range(len(pol)):
    G_phys = nx.barabasi_albert_graph(N, int(ave_degree/2))
    G_info_300 = G_phys.copy()
    utils.initNET_SI(G_info_300, I0=300, I_seed=np.arange(10))
    sim = SIR_net_adaptive(G_phys, G_info_300, beta=beta, mu=mu, r=r, pro=pro, pol=pol[i], p_sym=p_sym, initial_infecteds=initial_infecteds)
    networks.append(G_info_300)
    inf.append(sim[2])

simulation until time t=348

fig, ax = plt.subplots(nrows=5, ncols=4, figsize=(4*4, 3*3))
for i in range(len(pol)):
    utils.plot_info_network(networks[i])
    ax[i//5,i%4].plot(np.arange(len(inf[i])), inf[i], lw=3, c=colors(0), label= "0 No-Vax", alpha=0.6)
    

maxi = []
for i in inf:
    a = max(i)
    maxi.append(a)
    
plt.plot(pol,maxi)

[<matplotlib.lines.Line2D at 0x165bf563d30>]

A greater vaccination rate

N = 1000 # number of nodes
ave_degree = 15
p = ave_degree / N  # edge probability
initial_infecteds = np.random.choice(np.arange(N), size=100, replace=False)

mu = 0.02       # recovery rate
beta = 0.03    # infection rate
pro = 0.0005    # rate of classical media influence on people
pol = np.arange(1,20)/20       # propensity of opinion polarization
p_sym = 0.05    # prob for an infected person to have severe symptoms
r = 0.4         # rate of vaccination for PV

networks = []
inf = []

for i in range(len(pol)):
    G_phys = nx.barabasi_albert_graph(N, int(ave_degree/2))
    G_info_300 = G_phys.copy()
    utils.initNET_SI(G_info_300, I0=300, I_seed=np.arange(10))
    sim = SIR_net_adaptive(G_phys, G_info_300, beta=beta, mu=mu, r=r, pro=pro, pol=pol[i], p_sym=p_sym, initial_infecteds=initial_infecteds)
    networks.append(G_info_300)
    inf.append(sim[2])

simulation until time t=332

fig, ax = plt.subplots(nrows=5, ncols=4, figsize=(4*4, 3*3))
for i in range(len(pol)):
    utils.plot_info_network(networks[i])
    ax[i//5,i%4].plot(np.arange(len(inf[i])), inf[i], lw=3, c=colors(0), label= "0 No-Vax", alpha=0.6)

maxi = []
for i in inf:
    a = max(i)
    maxi.append(a)
    
plt.plot(pol,maxi)

[<matplotlib.lines.Line2D at 0x165a022c0d0>]